| vkoskiv | There are some places that can 3D print different metal alloys. | 00:07 |
|---|---|---|
| vkoskiv | Which sounds like a fun thing to try for the buttons there, perhaps | 00:08 |
| + bkeys (~Thunderbi@static-198-54-135-69.cust.tzulo.com) | 00:38 | |
| - bkeys (QUIT: Quit: With every step we take, danger will follow closely) (~Thunderbi@static-198-54-135-69.cust.tzulo.com) | 01:59 | |
| + bkeys (~Thunderbi@static-198-54-135-69.cust.tzulo.com) | 01:59 | |
| - mtm (QUIT: Ping timeout: 256 seconds) (~mtm@c-73-27-62-116.hsd1.fl.comcast.net) | 02:03 | |
| - bkeys (QUIT: Remote host closed the connection) (~Thunderbi@static-198-54-135-69.cust.tzulo.com) | 02:10 | |
| + bkeys (~Thunderbi@static-198-54-135-69.cust.tzulo.com) | 02:17 | |
| cinap_lenrek | hm | 02:30 |
| cinap_lenrek | getting erratic battery status readings | 02:30 |
| cinap_lenrek | i had the batteries disconnected for a while | 02:30 |
| cinap_lenrek | now, when the system is off, the right 4 cells show 3.3v | 02:31 |
| cinap_lenrek | the the last left hand side battery shows ~5.xv | 02:31 |
| cinap_lenrek | all other cells show 0.0v | 02:31 |
| cinap_lenrek | when i power the system up | 02:31 |
| cinap_lenrek | all but the bottom left hand side cell show 0.0v | 02:32 |
| cinap_lenrek | ok | 02:54 |
| cinap_lenrek | replugged batteries | 02:54 |
| cinap_lenrek | seems to be back to normal now | 02:54 |
| cinap_lenrek | all cells show voltage now | 02:54 |
| - bkeys (QUIT: Ping timeout: 276 seconds) (~Thunderbi@static-198-54-135-69.cust.tzulo.com) | 03:07 | |
| - nocko (QUIT: Ping timeout: 255 seconds) (sid501219@id-501219.helmsley.irccloud.com) | 03:41 | |
| - jfred (QUIT: Ping timeout: 255 seconds) (sid534649@libera/sponsor/jfred) | 03:45 | |
| + bkeys (~Thunderbi@static-198-54-135-69.cust.tzulo.com) | 04:00 | |
| + jfred (sid534649@libera/sponsor/jfred) | 04:01 | |
| + mtm (~mtm@c-73-27-62-116.hsd1.fl.comcast.net) | 04:09 | |
| + nocko (sid501219@user/nocko) | 04:13 | |
| - bkeys (QUIT: Ping timeout: 256 seconds) (~Thunderbi@static-198-54-135-69.cust.tzulo.com) | 04:20 | |
| + grubman (~ufotofu@149.248.16.17) | 05:06 | |
| grubman | hey all, what are the options for upgradeable CPU + RAM for the reform? i see on the home page that they are upgradeable. | 05:10 |
| grubman | i'd lke to at least bump the RAM up to 8 GB | 05:10 |
| - Christoph_ (QUIT: Remote host closed the connection) (~Christoph@p54bf604a.dip0.t-ipconnect.de) | 05:38 | |
| - Ar|stote|is (QUIT: Quit: https://quassel-irc.org - Chat comfortably. Anywhere.) (~linx@149-210-4-100.mobile.nym.cosmote.net) | 05:41 | |
| + Ar|stote|is (~linx@149-210-4-100.mobile.nym.cosmote.net) | 05:42 | |
| - jfred (QUIT: Ping timeout: 258 seconds) (sid534649@libera/sponsor/jfred) | 05:56 | |
| + jfred (sid534649@libera/sponsor/jfred) | 05:59 | |
| josch | grubman: right now there is only a single option. More options are in the works, namely a ls1028a based board https://source.mnt.re/reform/mnt-reform-layerscape-ls1028a-som and a board that makes the reform compatible with boards in the raspberry pi CM4 module form factor (including the rpi cmp4) | 06:37 |
| josch | or RK3566 based pine64 soquartz | 06:39 |
| kfx | rpi cm4 are tough to find these days | 06:39 |
| josch | kfx: do you find any other from scalpers? :D | 06:40 |
| josch | kfx: lukas puts higher priority on the ls1028a board which i like | 06:40 |
| kfx | it's the rpi foundation's own fault for prioritizing industrial customers | 06:40 |
| josch | * there is supposed to be a dash between "any" and "other" | 06:40 |
| kfx | yeah I'm far more interested in the ls1028a than in the rpi (no availability) or the rockhip (no drivers) | 06:41 |
| kfx | although to be honest now that there are 11,495 million chinese single-board computer vendors I am surprised at how few of them slid into the cm4 vacuum with compatible boards | 06:43 |
| josch | flaky or missing suspend currently makes the reform unfit for me as a daily driver -- i usually have 10 workspaces open with stuff on all of them. Is there hope for a new board to better support suspend? | 06:49 |
| - mlarkin (QUIT: Quit: Lost terminal) (~mlarkin@047-048-086-214.biz.spectrum.com) | 08:27 | |
| + MajorBiscuit (~MajorBisc@c-001-003-023.client.tudelft.eduvpn.nl) | 08:32 | |
| + mjw (~mark@gnu.wildebeest.org) | 09:29 | |
| + cinap_lenwerk (~cinap_len@196.red-88-12-2.staticip.rima-tde.net) | 10:37 | |
| cinap_lenwerk | my mnt reform shuts down after a few seconds when running off battery | 10:38 |
| cinap_lenwerk | no idea why | 10:38 |
| cinap_lenwerk | i had it plugged in the night on ac | 10:38 |
| cinap_lenwerk | so batteries should be charged | 10:39 |
| cinap_lenwerk | and when its off the voltages all show 3.3v on all cells | 10:39 |
| cinap_lenwerk | any tips? | 10:39 |
| - buckket (QUIT: Ping timeout: 248 seconds) (~buckket@pdp8.buckket.org) | 10:46 | |
| cinap_lenwerk | is it normal that th battery voltages change depending on if the imx8 is powered on/off? | 12:22 |
| + wielaard (~mjw_@2001:1c06:2488:1400:9e5c:8eff:fe8f:a440) | 13:08 | |
| - mjw (QUIT: Killed (NickServ (GHOST command used by wielaard!~mjw_@2001:1c06:2488:1400:9e5c:8eff:fe8f:a440))) (~mark@gnu.wildebeest.org) | 13:08 | |
| * wielaard -> mjw | 13:08 | |
| + mark__ (~mark@gnu.wildebeest.org) | 13:09 | |
| vkoskiv | I'm not an electrical engineer but I'd expect the current draw of the SoM to pull the voltage down somewhat. | 13:39 |
| + buckket (~buckket@pdp8.buckket.org) | 13:45 | |
| - XgF (QUIT: Remote host closed the connection) (~quassel@2001:19f0:5001:1174:5400:2ff:fef2:1fa2) | 13:49 | |
| + XgF (~quassel@2001:19f0:5001:1174:5400:2ff:fef2:1fa2) | 13:50 | |
| - mtm (QUIT: Ping timeout: 276 seconds) (~mtm@c-73-27-62-116.hsd1.fl.comcast.net) | 14:04 | |
| + Christoph_ (~Christoph@p4fe73ef6.dip0.t-ipconnect.de) | 14:10 | |
| + bkeys (~Thunderbi@static-198-54-135-69.cust.tzulo.com) | 14:14 | |
| josch | bkeys: there is a new sd-card image you can try out which includes the fixes for the prolbem nocko found: https://source.mnt.re/reform/reform-system-image/-/jobs/826/artifacts/browse/reform2-imx8mq/ | 14:35 |
| josch | bkeys: i tested both of these images on my sd-card successfully -- both boot fine over here | 14:35 |
| josch | vkoskiv: you can now "sudo apt install box64" -- i'd be happy about any feedback you can provide so that i can fix the packaging | 14:36 |
| + GNUmoon (~GNUmoon@gateway/tor-sasl/gnumoon) | 15:08 | |
| - GNUmoon (QUIT: Remote host closed the connection) (~GNUmoon@gateway/tor-sasl/gnumoon) | 15:08 | |
| + GNUmoon (~GNUmoon@gateway/tor-sasl/gnumoon) | 15:09 | |
| + chomwitt (~chomwitt@2a02:587:dc0d:e600:d03e:b48f:9497:fc81) | 15:12 | |
| flowy | josch: if you're adding that ytdl profile stuff to the mpv defaults in the sysimage skel, might consider also ignoring vp9, i.e.: | 15:54 |
| flowy | ytdl-format=bestvideo[height<=?1080][vcodec!=vp9]+bestaudio/best | 15:54 |
| flowy | i don't think the driver supports vp9 currently | 15:55 |
| flowy | i've personally given up with mpv... there's something about my setup that makes everything chunky despite hw accel. i just use clapper instead | 15:56 |
| flowy | might be my use of pipewire | 15:57 |
| + mtm (~mtm@c-73-27-62-116.hsd1.fl.comcast.net) | 16:09 | |
| josch | flowy: then you probably should also ignore av01, no? | 16:14 |
| - bkeys (QUIT: Ping timeout: 255 seconds) (~Thunderbi@static-198-54-135-69.cust.tzulo.com) | 16:18 | |
| + bkeys (~Thunderbi@static-198-54-135-69.cust.tzulo.com) | 16:25 | |
| flowy | i guess yeah! | 16:36 |
| josch | and that'd make me worry whether this will disable video completely on some platforms supportey by youtube-dl? | 16:37 |
| flowy | right... oh well. just seems a bit funny, if people are using this mostly for youtube, to put hwdec=auto in there without ignoring the codecs that ytdl will fetch from youtube 99% of the time that will make hwdec not possible. | 16:44 |
| flowy | though i guess based on my own experience hwdec doesn't even work very well | 16:45 |
| flowy | CPU decoding and limiting to practical resoltutions is effectively the better default | 16:46 |
| josch | flowy: i think instead of forbidding vp9 it would be better to adjust the format sorting via --format-sort | 16:48 |
| flowy | i did some tests again because i was curious. i took 3 random videos from youtube, dl'd best h264 @1080, ran mpv with audio disabled to make sure my pipewire setup isn't doing anything weird | 16:49 |
| flowy | hwdec=no played all of them smoothly | 16:49 |
| flowy | hwdec=auto made all of them chunky | 16:49 |
| flowy | confirmed the codec was being used monitoring interrupts | 16:49 |
| flowy | so yeah i'm just not using mpv w hw accel | 16:50 |
| flowy | clapper works great though | 16:50 |
| minute | is clapper already packaged? | 16:52 |
| + vagrantc (~vagrant@2600:3c01:e000:21:7:77:0:20) | 16:53 | |
| flowy | the flatpak @flathub works perfectly, no tweaks necessary to start using hw accel | 16:53 |
| flowy | on sysimagev3 i mean | 16:53 |
| minute | nice | 16:56 |
| flowy | i've actually been using clapper to watch a lot of stuff on the reform over the last few months | 16:56 |
| flowy | even 1080p60fps gaming streams work well | 16:56 |
| minute | nice | 16:58 |
| - xktr (QUIT: Remote host closed the connection) (~xktr@2a0a-e5c0-2-12-0-f0ff-fea9-c3dc.loves.ipv6.at.ungleich.ch) | 16:59 | |
| flowy | that's where you can basically never rely on mpv working without hw accel | 16:59 |
| + xktr (~xktr@2a0a-e5c0-2-12-0-f0ff-fea9-c3dc.loves.ipv6.at.ungleich.ch) | 17:01 | |
| - GNUmoon (QUIT: Remote host closed the connection) (~GNUmoon@gateway/tor-sasl/gnumoon) | 17:12 | |
| + GNUmoon (~GNUmoon@gateway/tor-sasl/gnumoon) | 17:12 | |
| - cinap_lenwerk (QUIT: Ping timeout: 246 seconds) (~cinap_len@196.red-88-12-2.staticip.rima-tde.net) | 17:25 | |
| + mlarkin (~mlarkin@047-048-086-214.biz.spectrum.com) | 17:42 | |
| + cinap_lenwerk (~cinap_len@ns3076381.ip-147-135-136.eu) | 17:47 | |
| cinap_lenrek | its really wired | 17:49 |
| cinap_lenrek | the issues with shutting down went away after resetting the lpc | 17:50 |
| cinap_lenrek | tho the battery status is still strange | 17:50 |
| cinap_lenrek | one cell now shows 0.0v | 17:50 |
| cinap_lenrek | other shows 5v sometimes | 17:50 |
| cinap_lenrek | other 0.9v | 17:50 |
| cinap_lenrek | like wtf | 17:51 |
| cinap_lenrek | wer misst misst mist | 17:51 |
| minute | maybe some of the cells are really busted? | 17:52 |
| minute | but weird @ reset | 17:52 |
| cinap_lenrek | all i did was having the cells disconnected for a while | 17:54 |
| cinap_lenrek | i removed the cables from the board | 17:54 |
| cinap_lenrek | while i was testing in the rack | 17:54 |
| cinap_lenrek | and they where nicely charged before | 17:54 |
| cinap_lenrek | i think it was around 4 weeks | 17:54 |
| minute | ah | 17:54 |
| minute | maybe the battery board connector is not all the way in at the motherboard? | 17:55 |
| minute | it has to click in | 17:55 |
| cinap_lenrek | you think its just a bad connection? | 17:55 |
| minute | if it is loose then those effects could happen yeah | 17:55 |
| cinap_lenrek | indeed | 17:55 |
| cinap_lenrek | system status says cell missing | 17:55 |
| cinap_lenrek | jesus | 17:58 |
| cinap_lenrek | fixed | 17:58 |
| cinap_lenrek | i'm so dumb | 17:58 |
| minute | what happened? | 17:59 |
| cinap_lenrek | it was like you said | 17:59 |
| minute | ah nice | 17:59 |
| cinap_lenrek | i didnt push in the cable fully | 17:59 |
| minute | well i'm very glad | 18:00 |
| cinap_lenrek | i'm such a desaster with hardware | 18:00 |
| cinap_lenrek | thank you man! | 18:00 |
| minute | well, you know how to make it work on the driver level | 18:00 |
| - MajorBiscuit (QUIT: Quit: WeeChat 3.5) (~MajorBisc@c-001-003-023.client.tudelft.eduvpn.nl) | 18:15 | |
| kfx | I demand to see cinap's battery cable drivers | 18:16 |
| - bkeys (QUIT: Ping timeout: 240 seconds) (~Thunderbi@static-198-54-135-69.cust.tzulo.com) | 18:59 | |
| vkoskiv | I just ran apt upgrade and sway broke | 19:26 |
| minute | vkoskiv: broke in what way? | 19:26 |
| vkoskiv | Failed to initialize EGL context | 19:26 |
| minute | huh | 19:26 |
| minute | which version of sway is that? | 19:26 |
| vkoskiv | 00:00:00.181 [wlr] [EGL] command: eglQueryDeviceStringEXT, error: EGL_BAD_PARAMETER (0x300c), message: "eglQueryDeviceStringEXT" | 19:27 |
| vkoskiv | 00:00:00.181 [wlr] [EGL] command: eglQueryDeviceStringEXT, error: EGL_BAD_PARAMETER (0x300c), message: "eglQueryDeviceStringEXT" | 19:27 |
| vkoskiv | 00:00:01.936 [wlr] [render/egl.c:291] Software rendering detected, please use the WLR_RENDERER_ALLOW_SOFTWARE environment variable to proceed | 19:27 |
| vkoskiv | 00:00:01.937 [wlr] [render/egl.c:521] Failed to initialize EGL context | 19:27 |
| vkoskiv | 00:00:01.937 [wlr] [render/gles2/renderer.c:698] Could not initialize EGL | 19:27 |
| vkoskiv | 00:00:02.221 [wlr] [backend/libinput/backend.c:116] libinput initialization failed, no input devices | 19:27 |
| vkoskiv | 00:00:02.221 [wlr] [backend/libinput/backend.c:117] Set WLR_LIBINPUT_NO_DEVICES=1 to suppress this check | 19:27 |
| vkoskiv | 00:00:02.221 [wlr] [backend/multi/backend.c:34] Failed to initialize backend. | 19:27 |
| vkoskiv | 00:00:02.222 [sway/server.c:293] Failed to start backend | 19:27 |
| minute | well, what is the sway version? | 19:27 |
| vkoskiv | vkoskiv@reform:~$ sway | 19:27 |
| vkoskiv | vkoskiv@reform:~$ /server.c:293] Failed to start backendo initialize backend.ES=1 to suppress this checkTWARE environment variable to proceed | 19:27 |
| vkoskiv | v1.7 | 19:27 |
| minute | i'm running the same without a problem | 19:27 |
| vkoskiv | Hmm. Let me try another reboot. | 19:28 |
| minute | did you create a new user account without privileges or something? | 19:28 |
| minute | i do run a custom mesa though, i should probably nuke it | 19:30 |
| vkoskiv | I didn't do anything other than a apt upgrade + a reboot | 19:30 |
| minute | ok, let me clean out my /usr/local/lib | 19:32 |
| flowy | i'm on sysimagev3, also apt upgraded just a couple hours ago and sway is fine for me | 19:32 |
| - mjw (QUIT: Quit: Leaving) (~mjw_@2001:1c06:2488:1400:9e5c:8eff:fe8f:a440) | 19:32 | |
| flowy | oh but wait a min | 19:33 |
| flowy | i vaguely remember the error you're getting from an upgrade months ago | 19:33 |
| vkoskiv | Nothing weird shows up on boot | 19:33 |
| minute | works for me even after cleaning out /usr/local/lib | 19:33 |
| minute | and i'm on latest mesa from debian | 19:33 |
| vkoskiv | Weird. I'll try reinstalling sway? | 19:33 |
| flowy | it's input related or something | 19:33 |
| vkoskiv | No updates available on sway | 19:33 |
| flowy | i ran sway through strace to figure it out | 19:34 |
| vkoskiv | s/sway/apt | 19:34 |
| minute | vkoskiv: do you have anything in /usr/local/lib ? | 19:34 |
| flowy | i had the exact same error | 19:34 |
| vkoskiv | I have python 3.10 and 3.9 in there | 19:34 |
| minute | interestingly kmscube doesn't work for me | 19:34 |
| minute | ah it does | 19:35 |
| minute | vkoskiv: in a console, can you run kmscube -D /dev/dri/card0 | 19:35 |
| minute | josch: wow! after exiting "kmscube -D /dev/dri/card0", my console is gone! | 19:35 |
| minute | ah, but alt+f2 brings it back | 19:36 |
| minute | maybe that would also work in the case of no display at boot... | 19:36 |
| vkoskiv | I get "failed to set mode: invalid argument" | 19:36 |
| vkoskiv | And a bunch of opengl info before that | 19:36 |
| minute | vkoskiv: hmmm maybe libdrm messed up... or kernel | 19:37 |
| minute | vkoskiv: do you have /dev/dri/renderD128? | 19:37 |
| vkoskiv | It did upgrade some kernel/dts stuff | 19:37 |
| flowy | i'm going through my bash history and stuff to try to remember how i fixed it | 19:37 |
| vkoskiv | minute: I don't have that, no. | 19:37 |
| minute | vkoskiv: that means the GPU module was not loaded | 19:37 |
| minute | mega weird | 19:38 |
| vkoskiv | Just card0 in there | 19:38 |
| minute | vkoskiv: ok, so your etnaviv is gone | 19:38 |
| vkoskiv | Huh. Does apt log install output anywhere? | 19:38 |
| vkoskiv | I could share that | 19:38 |
| minute | not needed | 19:38 |
| minute | vkoskiv: sudo lsmod | grep etna | 19:39 |
| minute | vkoskiv: also, sudo dmesg | grep etna | 19:39 |
| vkoskiv | Nothing | 19:39 |
| vkoskiv | Nothing | 19:39 |
| minute | wow | 19:39 |
| minute | vkoskiv: sudo modprobe etnaviv | 19:39 |
| vkoskiv | You're gonna love this | 19:40 |
| vkoskiv | [ 663.706324] gpu_sched: disagrees about version of symbol module_layout | 19:40 |
| vkoskiv | modprobe: ERROR: could not insert 'etnaviv': Exec format error | 19:40 |
| minute | wow | 19:40 |
| vkoskiv | Say what now? :D | 19:40 |
| vkoskiv | I have an idea | 19:40 |
| vkoskiv | I get an error on boot about binfmt formats | 19:40 |
| vkoskiv | Maybe, just maybe it's confused because of that? | 19:40 |
| minute | maybe it tried to load x86 modules or something weird? | 19:40 |
| minute | did you mess with architectures? | 19:41 |
| vkoskiv | No, I ended up removing amd64 after some messing around | 19:41 |
| vkoskiv | So no amd64 packages installed | 19:41 |
| minute | aha, but you did mess with architectures | 19:41 |
| vkoskiv | But my binfmt things are still in there | 19:41 |
| minute | you had it installed and then removed | 19:41 |
| vkoskiv | Yeah | 19:41 |
| minute | ok, try to clean that up | 19:42 |
| vkoskiv | Sure, sec. | 19:42 |
| minute | i'm also upgrading to 5.18 from our repo now | 19:43 |
| josch | i tested the sysimage-v3 that was built today and thus should also have the latest package versions and that one boots up fine including sway, gnome and plasma | 19:44 |
| flowy | vkoskiv: i've also seen that module error. in my case it was because i ran apt upgrade at a time when my /boot partition was unavailable | 19:45 |
| flowy | which caused a situation where i was running a different version kernel than modules that were trying to load | 19:46 |
| vkoskiv | Cleaned up binfmt stuff, still no sway | 19:46 |
| minute | this sounds likely flowy -> vkoskiv | 19:46 |
| vkoskiv | flowy: My /boot comes from emmc, and I haven't unmounted it | 19:46 |
| minute | vkoskiv: do you have /lib/modules/5.18.0-reform2-arm64? | 19:47 |
| vkoskiv | Is there a way to verify that theory? | 19:47 |
| vkoskiv | minute: I do | 19:47 |
| minute | vkoskiv: what is `file /lib/modules/5.18.0-reform2-arm64/kernel/drivers/gpu/drm/etnaviv/etnaviv.ko` ? | 19:48 |
| minute | vkoskiv: and md5sum on that file? | 19:48 |
| minute | should be a6fdedf679d... | 19:48 |
| vkoskiv | /lib/modules/5.18.0-reform2-arm64/kernel/drivers/gpu/drm/etnaviv/etnaviv.ko: ELF 64-bit LSB relocatable, ARM aarch64, version 1 (SYSV), BuildID[sha1]=5974ae942e010b7deb08858b5015d368e3ec4cd1, not stripped | 19:48 |
| vkoskiv | Wonder where that came from, then | 19:48 |
| minute | ah yeah, if i try to insmod this, i get the same error | 19:49 |
| minute | because i'm running kernel 5.17 | 19:49 |
| minute | (i haven't rebooted yet) | 19:49 |
| flowy | so what's your boot layout right now | 19:49 |
| flowy | and fstab | 19:49 |
| minute | so it's a kernel version vs module version mismatch vkoskiv | 19:49 |
| minute | vkoskiv: what's `uname -a`? | 19:49 |
| vkoskiv | 5.18.2 kernel | 19:50 |
| minute | 5.18.2?? | 19:50 |
| vkoskiv | -reform2-arm64 yadda yadda | 19:50 |
| minute | not 5.18.0? | 19:50 |
| vkoskiv | Linux reform 5.18.0-reform2-arm64 #1 SMP Debian 5.18.2-1+reform1 (2022-06-08) aarch64 GNU/Linux | 19:50 |
| vkoskiv | My bad | 19:50 |
| vkoskiv | I was looking at that Debian part there | 19:50 |
| flowy | yeah but that's not up to date | 19:50 |
| flowy | so your modules got upgraded | 19:50 |
| vkoskiv | I just upgraded it | 19:51 |
| flowy | your kernel didn't | 19:51 |
| flowy | yeah but it didn't work for some reason | 19:51 |
| vkoskiv | Huh. How do I retry the kernel update? | 19:51 |
| flowy | i wonder if you're not reaching the right /boot | 19:51 |
| vkoskiv | It's been the emmc one this whole time, haven't messed with that | 19:51 |
| flowy | but is your root filesystem somewhere else? | 19:51 |
| vkoskiv | My uboot is on the sd card thoug | 19:51 |
| flowy | your active one | 19:51 |
| vkoskiv | Yeah the root one is on nvme | 19:51 |
| vkoskiv | encrypted lvm | 19:51 |
| minute | vkoskiv: what's `mount | grep boot` ? | 19:51 |
| vkoskiv | /dev/mmcblk0p1 on /boot type ext2 (rw,relatime,errors=remount-ro) | 19:52 |
| minute | that seems correct | 19:52 |
| minute | vkoskiv: an sd card is not inserted? | 19:52 |
| vkoskiv | It is inserted | 19:52 |
| minute | what is on the sd card? | 19:52 |
| vkoskiv | But not mounted according to lsblk | 19:52 |
| vkoskiv | It have uboot on there I think, won't boot without it | 19:52 |
| vkoskiv | I didn't get the no-sd card boot working yet | 19:53 |
| minute | ah. | 19:53 |
| minute | so u-boot will probably load the kernel from that sd card | 19:53 |
| minute | and not from emmc | 19:53 |
| vkoskiv | Yeah, lsblk shows sd card but no mounted partitions | 19:53 |
| minute | if you boot from sd card, your /boot needs to be on the sd card AFAIK | 19:53 |
| vkoskiv | And this just happened to be the first kernel+module upgrade since? That makes sense | 19:54 |
| vkoskiv | Ah. So I need to fix this to boot from emmc then :D | 19:54 |
| minute | so, i would suggest you change your fstab | 19:54 |
| minute | no, first fix the boot from sd card | 19:54 |
| vkoskiv | Okay | 19:54 |
| flowy | yeah i forget how i fixed my mismatch... i think i somehow told apt to reinstall the latest kernel after pointing /boot to the right location again | 19:55 |
| minute | edit /etc/fstab and change mmcblk0p1 to mmcblk1p1 | 19:55 |
| josch | alternatively, delete boot.scr from the first partition on your sd-card, then u-boot will fall back to loading boot.scr (and kernel+initramfs) from emmc where you have the right versions | 19:55 |
| minute | ah | 19:55 |
| minute | ok, i did not know this was possible | 19:55 |
| vkoskiv | I seem to remember emmc boot never working before | 19:55 |
| vkoskiv | I didn't get it quite figured out, didn't know what the issue was with that | 19:56 |
| vkoskiv | I tried every permutation of the boot config and switch, and it never booted without and sd card in | 19:56 |
| flowy | if you unmount /boot, change fstab, remount /boot, then run 'apt-get install --reinstall <yer latest kernel pkg>', that should also work | 19:57 |
| vkoskiv | I'll try that | 19:58 |
| flowy | get the latest pkg name: 'dpkg -l|grep linux-image' | 19:58 |
| vkoskiv | So linux-image-5.18.0-reform2-arm64? | 19:59 |
| flowy | yeah | 19:59 |
| flowy | heh, yeah, this is actually what happened to me months ago when sway wouldn't start | 20:00 |
| flowy | it took me a while to figure out it had nothing to do with libinput blah blah | 20:00 |
| flowy | and that my kernel didn't upgrade properly | 20:00 |
| minute | sounds like we will need some kind of integrity checking tool ^^ | 20:01 |
| flowy | my setup is just a little different in that my root is on nvme but uboot and /boot are on a little 1GB pslc swissbit card | 20:02 |
| flowy | and yeah it's easy for the sd card to get pushed out without noticing | 20:02 |
| flowy | so i had upgraded at a time when /boot wasn't available | 20:02 |
| flowy | and not noticed | 20:02 |
| flowy | then rebooted and things were out of sync | 20:02 |
| flowy | in your case i guess there weren't any errors because /boot is available but it's the wrong /boot | 20:03 |
| flowy | but my sd card doesn't contain a linux rootfs or anything, just /boot | 20:04 |
| vkoskiv | I did the fstab boot partition upgrade trick, and now it works again! | 20:05 |
| flowy | yay | 20:05 |
| vkoskiv | And I learned a bit. I think I know how to fix this, should it happen again | 20:05 |
| vkoskiv | Thanks for the help! | 20:05 |
| minute | josch: i think your USB-C charger is really only for iphone and galaxy, and not a USB-C PD charger | 20:09 |
| minute | it says something like 20W on the amazon page but reform charges with almost 60W and at 24V (or 20V in the case of USB-C PD) | 20:10 |
| minute | vkoskiv: great | 20:11 |
| flowy | usb c power brought back plug & pray | 20:17 |
| vkoskiv | The true wild card of connectors | 20:17 |
| flowy | sometimes you can't even be certain of which device is going to supply the other power! | 20:18 |
| * mark__ -> mjw | 20:21 | |
| + bkeys (~Thunderbi@static-198-54-135-69.cust.tzulo.com) | 21:23 | |
| bkeys | josch: Okay I am about to try out that image on my Reform now that I am free | 21:24 |
| bkeys | Once I confirm a working OS I am probably gonna get it ready to ship to the people who will hopefully repair it | 21:24 |
| josch | minute: interesting, so even though I was able to charge my reform with a 10W power supply, I need a 60 W power supply if I want to do usb-c charging? :( | 21:29 |
| josch | vkoskiv: glad to hear that it works for you again -- this would all be easier if uboot were able to turn on the internal display and then you would see what is happening :/ | 21:30 |
| josch | bkeys: much success! I'm afraid I'll only be back online in about 12 hours but I'll read your messages in case you write more tonight. :) | 21:31 |
| minute | josch: no, it does not need to have 60W. but you need more than 5V | 21:36 |
| minute | josch: i'm not sure how the apple+galaxy charging works, but it's not the normal USB-C PD charging | 21:36 |
| josch | oh no :/ | 21:36 |
| minute | afaik apple just dumps more power at 5v than usb normally allows? | 21:37 |
| josch | I followed this heise review www.techstage.de/nz3jr2s and it claimed that the "Anker PowerPort III Nano" could do power delivery -- apparently PD is not equal to PD :/ | 21:37 |
| minute | ah, quick charge can go higher but i guess it's negotiated in a different way that usb-c pd | 21:37 |
| josch | On the german amazon page for that product it also says nothing about PD but on the english page it says "and supports USB-C Power Delivery" which was what made me decide for this one. | 21:38 |
| minute | huh. | 21:38 |
| josch | yeah, very odd | 21:39 |
| minute | kind of unclear to me from anker's website | 21:42 |
| minute | i don't see the PD logo anywhere | 21:42 |
| minute | but sometimes it's mentioned, sometimes not | 21:43 |
| sknebel | "PD" also doesnt mean it can do all the voltages | 21:43 |
| sknebel | i.e. it might be able to negotiate up to 12V, or even just 5V with high current | 21:43 |
| sknebel | but not 20V | 21:43 |
| josch | puh, okay, thanks for your input. I'm going on holidays for 10 days starting on Thursday and i'll take the reform with me as my only laptop as its first battle test (no worries, i'll also take a screwdriver with me) and my hope was that i could just share an AC adapter with my wife but i have another very small AC adapter that i'll take instead. | 21:44 |
| - GNUmoon (QUIT: Ping timeout: 268 seconds) (~GNUmoon@gateway/tor-sasl/gnumoon) | 21:44 | |
| minute | ah yeah i just wanted to write what sknebel just wrote | 21:45 |
| josch | sknebel: what is the thing that I have to look out for when buying a usb-c ac adapter if just PD is not enough? | 21:45 |
| minute | https://en.wikipedia.org/wiki/USB_hardware#USB_Power_Delivery | 21:45 |
| josch | and yes, I now see that the anker website doesn't explicitly list PD for this device -- so the techstage.de information is just wrong... | 21:45 |
| sknebel | 45W or above afaik is always 20V | 21:45 |
| minute | yeah, i think the wattage will imply it | 21:46 |
| sknebel | basically usually it'll go up to 3 amps | 21:46 |
| sknebel | and voltage levels are 5,9,12,15,20 | 21:47 |
| josch | okay, sounds like choosing a 60 W minimum usb-c AC adapter will do the trick for me -- i'll try that out in two weeks then. Thank you guys! :) | 21:47 |
| minute | yes, that should work | 21:47 |
| sknebel | so I guess 45W could in theory also be 15V at 3A. but 60W will have to be 20 | 21:47 |
| minute | yep | 21:48 |
| sknebel | (over 60W also exists, but is then special case with supported cables that signal "I wont go up in flames if you push more through me") | 21:48 |
| sknebel | (and then there is USB PD PPS, where the voltage can be freely choosen, ... its complicated :D) | 21:49 |
| + GNUmoon (~GNUmoon@gateway/tor-sasl/gnumoon) | 21:57 | |
| minute | oof, didn't know about PPS | 22:24 |
| vkoskiv | josch: Enjoy the holiday! | 22:29 |
| kfx | the anker powerpoint iii nano only goes up to 9v@2.2A (I have one and have tested it) | 22:33 |
| kfx | I have successfully charged the reform with both of these devices https://wiki.pine64.org/wiki/PinePower and a usb-c - type k barrel cable which negotiates 20V | 22:35 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!